[Help Thread] FMC Discussion

[irontwig]

Pardon my denseness, but what is an example of this?

U M U2 M' U

 

[Mike Hughey]

U M U2 M' U

Oh, that one! Okay, thanks.

 

[Cubenovice]

I think this solve (example solve thread #1179) of Irontwig deserves to be copied here:
All conjugates in blue...

scramble: F L2 D2 U2 B' R2 L2 U' F D L' U2 D F' L2 R' B D2 B D' F L' D2 F L

U' L' U . D' B D L2
L F2 L' F2 U' L R' F2 L' R2 U2 : R2 F R F' U
R2

.=U' L U R2 U' L' U R2 cancels 6 moves
:=U2 F' D2 F U2 F' D2 F

 

[JackJ]

Critique please. This is a really good and lucky FMC solution for me, 42 moves, one rotation.

U2 D F' B' U' R2 L U F2 D L D' L2 B2 D2 U' F2 B' L' U R2 F U' L' F2

z2 R U R' B U D' R D2
U L F' U2 L U' B' U B
U' F' U F L F' L' F
U2 F' U' F R' F R
F R2 F L F' R2 F L' F

 

[Cubenovice]

Critique please. This is a really good and lucky FMC solution for me, 42 moves, one rotation.

A quick check of your start:
Right at the start of your 8 move 2x2x2 you break up a nice “square” (center-edge-center)
This is a big no-no ;-)
Try to find a way to build a new square that you can align with this one to form a (pseudo) 2x2x2

Try z2 U2 B2 U for a 3-move 2x2x2 block (OK, it is a pseudo making it 4 moves if you like)
L’ B D’ another square
L’ U B’ place pair next to square to form 1x2x3 block
U’ B2 U’ move block out of the way, place edge, align block
This is 12 moves for a pseudo F2l-slot (premove D2 to make this a regular F2L-slot in 13 moves)

And this premove is again the reason why you should never, never write with cube rotations.
The premove to apply before the scramble is U2!!!

 

[kinch2002]

Stuck after a really nice start. Anyone fancy helping?
[23:49] <+statue> ,3fmc
[23:49] <+Nibblr> 3x3 Scramble #20125: R L' B2 L' D' L2 U2 F L2 F L2 D2 L' D2 B' U' B L' D2 B D B U B R2
Normal Scramble
2x2x3: U B U' B2 D
Inverse scramble with PM: F' D' B2 U B' U' (6)
F2L: F2 L F' L2 F' L' F2 (13)
Anyone got a good ending/adaptation?
Also, if anyone's interested on the normal scramble you can do U B U' B2 L2 D (6) for EO as well as the block

 

[Cubenovice]

Stuck after a really nice start. Anyone fancy helping?
<+Nibblr> 3x3 Scramble #20125: R L' B2 L' D' L2 U2 F L2 F L2 D2 L' D2 B' U' B L' D2 B D B U B R2

Also, if anyone's interested on the normal scramble you can do U B U' B2 L2 D (6) for EO as well as the block

normal scramble U B U' B2 L2 D (6) for EO as well as the block
L F' L2
D' F' D F' D' F2 D F2 Sune leaves 5 corners after 17 HTM
should be sub 30 with insertions

Edit: ehm yeah.... leaves 6 corners... :fp

 

[DavidWoner]

Here's some movecounts that I gathered from experience, imo it's very useful to know what's a good start and how many moves you can expect to finish up in. Feel free to correct and such.

Some movecounts for good starts:

Square+pair: 3
2x2x2: 4
Two squares: 5
2x2x3: 9
F2L-1: 14
F2L: 17

I don't usually keep track of my square starts, but my experience agrees with the other numbers.

 

[manyhobbyfreak]

just got really nice solve. (for me atleast)
it leaves 3 edges and is 22 moves so far

Scramble: L2 B F L R' D' U B' L' D R2 D' F' L2 B2 D2 F' L2 U2 B' D L U2 F' U

do the inverse
2X2 block: D' B2 R B2 4/4
solving rest of F2l + oll skip: F' U' F2 U' F2 R U' R' F2 L2 U L2 F L' F' U2 F U2: 18/22

does anyone find a nice place to solve the edges?

 

[ilikecubing]

Scramble : B' L2 D R2 U' D B2 L D2 F R' F2 D' R2 B' D2 F2 B2 R' B2 R F2 U R2 D

2x2x2: F2 D2 F U
2x2x3: D' F D' R D B' D B
F2L3: L D L' F D F' D' R F' R'
F2L4 + WV: D L D L' D' L D2 L' R' D2 L D L' D2 R
PLL: F D2 F' D' F D2 B' D F' D' B

48 moves

I felt like I had a really good start when I got the 2x2x2 so easily,but then it all went wrong,I'm sure I took more than half an hour looking for a low move extension to 2x2x3 to continue the good start,but I failed and my morale went down and the rest of the solve was just LOL.anyway I'm a big FMC nub so any tips on how to improve will be great.

I'm trying to learn premoves but I am stuck at one place,how to find the premoves and basically what is the concept of premoves? How does it give a better start?

 

[okayama]

Scramble : B' L2 D R2 U' D B2 L D2 F R' F2 D' R2 B' D2 F2 B2 R' B2 R F2 U R2 D

2x2x2: F2 D2 F U
2x2x3: D' F D' R D B' D B
F2L3: L D L' F D F' D' R F' R'
F2L4 + WV: D L D L' D' L D2 L' R' D2 L D L' D2 R
PLL: F D2 F' D' F D2 B' D F' D' B

48 moves

I felt like I had a really good start when I got the 2x2x2 so easily,but then it all went wrong,I'm sure I took more than half an hour looking for a low move extension to 2x2x3 to continue the good start,but I failed and my morale went down and the rest of the solve was just LOL.anyway I'm a big FMC nub so any tips on how to improve will be great.

I'm trying to learn premoves but I am stuck at one place,how to find the premoves and basically what is the concept of premoves? How does it give a better start?

Hi. Here is my 15 min solution with explanation how I obtained:

I started with your solve,

c/e pair: F2 D2

and found an easy way to create another block:

2x2x1 block: R U F'
2x2x2 block: D R'

Then create more squares:

Two squares: L2 B U
Pseudo F2L minus 1 slot: B2 L

Pseudoness may be not so easy to recognize for you, so consider the following premoves to compensate it.

Turn premoves: B L
Then turn the scramble: B' L2 D R2 U' D B2 L D2 F R' F2 D' R2 B' D2 F2 B2 R' B2 R F2 U R2 D

Then again

c/e pair: F2 D2
2x2x1 block: R U F'
2x2x2 block: D R'
Two squares: L2 B U
F2L minus 1 slot: B2 L

You can find a clean F2L minus 1 slot this time.
Then proceed:

Orient edges: U2 L U' L'
Finish F2L: U' B' U B
LL: B U2 B' U' B U' B'

A final solution for the given scramble is obtained by adding premoves at the end, i.e.,

F2 D2 / R U F' D R' / L2 B U / B2 L / U2 L U' L' / U' B' U B / B U2 B' U' B U' B' / B L

Solution: F2 D2 R U F' D R' L2 B U B2 L U2 L U' L' U' B' U B2 U2 B' U' B U' L (26 HTM)

This is how premoves work. Premoves "B L" mean that

Consider the state obtained by moving L' B' from the solved state.
In other words, you can solve the cube by moving B L after that.
I recognize such a state as a final goal to reach.

If you turn my solution without premoves, you may find what I say.

 

[Cubenovice]

just got really nice solve. (for me atleast)
it leaves 3 edges and is 22 moves so far

Scramble: L2 B F L R' D' U B' L' D R2 D' F' L2 B2 D2 F' L2 U2 B' D L U2 F' U

do the inverse
2X2 block: D' B2 R B2 4/4
solving rest of F2l + oll skip: F' U' F2 U' F2 R U' R' F2 L2 U L2 F L' F' U2 F U2: 18/22

does anyone find a nice place to solve the edges?

Hi Tristan, nice to see that you are still trying FMC.
I hope my explanation (although a bit short in time) of corner cycle insertion at Nemo 2011 has been of some use.

Same concept applies to edges but unfortunately you'll typically find less opportunities for good insertions.
The ideal insertion is an edge cycle on a center slice, taking only 6 HTM for U2 M U2 M' style cycles
No such luck in your solve...

But you do have your three edges in an U-perm somewhere in your solve.
Optimal U-perm is 9 HTM (a= B2 U M U2 M' U B2 , b=B2 U' M U2 M' U' B2)
Apply stickers 1-2-3 to your three edges and see if you can find it.

With the 9- mover you're at 31 HTM for the full solve.

Cheers,
Ralph

NB: be very carefull with slice moves in FMC! Keep a close eye on those centers!
U2 M U2 M' is actually U2 L' R B2 L R'

 

[Sebastien]

I'd like to share my funny solution from Düsseldorf Open 2011

Scramble: D' R2 U R2 F2 D' B2 U' R2 F2 L2 U F' D' L F R U2 F' R D2 R

This is what I found within 15 minutes:

Pseudo 2x2x3: B' F D' F R' F' U R U' F' U2
Pseudo-Tripod: R B2 R U R' . U' B U R2 B R' B'
3-cycle: R2 B R B'

Just a 3-cycle in 27 moves, so nothing special at all. I still decided to make an insertion to have a decent safety solution. Then out of nowhere:

. = R U' L U R' U' L' U

6!!! move cancelation I have never ever had more than 5 move cancelation before!

Solution: B' F D' F R' F' U R U' F' U2 R B2 R L U R' U' L' B U R2 B R' B' R2 B R B'

29 moves, 27 moves for the 3-cycle but still sub30

 

[Julian]

Decided to do my first ever FMC Of course it's terrible, but I'm still familiarizing myself with the techniques.

Scramble: L F' B' R' D2 F R F U2 R' F' L2 U' F' D' R' D' B' R' F' L' F' U B' L2

2x2x3: D' B' D2 B2 D' . B' U F B2 R U
Cross + C of F2L3 + E of F2L4: F R F' R' F
Finish F2L: U L D2 L' D L .. D' L' U'
OLL: F' R' D' R D F
Adjust: D2

Insert edge 3-cycle YB->YR->YO (F' B U2 F B' R2) at ".", cancelling no moves.
Insert corner 3-cycle YRB->YGR->YOG (R' U R D' R' U' R D) at "..", cancelling 2 moves.

Final Solution:

Spoiler

D' B' D2 B2 D' F' B U2 F B' R2 B' U F B2 R U
F R F' R' F
U L D2 L' D L R' U R D' R' U' R L' U'
F' R' D' R D F
D2

44 moves.

 

[Rpotts]

that's an interesting solution, when I read your solution and say the Cross+ 1 edge and one corner thing then Finish F2L I kinda facepalmed, but you actually did that really well. Not exactly the best FMC technique but 44 moves for first try isn't bad at all. I would recommend creating a F2L-1 then manipulating the last layer as much as you can before finishing the F2L (if you do at all,) like make a tripod or orient edges or reduce to L3C. I'm going to see if I can't find a better continuation after your 2x2x3.


Scramble: L F' B' R' D2 F R F U2 R' F' L2 U' F' D' R' D' B' R' F' L' F' U B' L2

2x2x3: D' B' D2 B2 D' B' U F B2 R U

F2L-1: F D U L D' L' U'

Tripod: D2 F D' F'


ugh that leaves a terrible terrible continuation.

I may try again later.

 

[Brest]

OLL: F' R' D' R D F

The B/F mirror of this alg still achieves OLL; however you will leave 3 corners instead of a G perm.

B R D R' D' B'

Mirroring algs is very important in FMC, they are kind of 'free' and give you many more options.

It is also important to not look at solving the last layer the same as you would in a speed solve. For example: OLL then PLL is not very move efficient (although some times it can be). Your goal at this stage is not orienting the LL, it's leaving a 3 cycle. (Well, really it's solving the cube :rolleyes: but a short LL is usually just luck.) Try block building from "F2L-1 slot" so that when you finish the F2L some of the LL is also complete, be it just edge orientation, a CE pair, or a 1x2x2 block.

In your example the OLL used orients the edges (well, obviously) but it left 3 corners and 3 edges. When this happens you could try a different OLL that also orients the edges and maybe find one that completely solves the edges and also leaves 3 corners. This was a relatively lucky case where the mirror does just that, but there are other short LL algs that can be used to fix EO and one of them just might leave a 3 cycle.

 

[Jorghi]

O_O 2x2x3??? Is that what FMC is??? How long does it take to plan those?

Wouldn't it be good to do 2x2x2 blocks in normal 3x3x3 solving. Sort of like making a reduction method? Because its fast in 4x4x4.

 

[Julian]

The B/F mirror of this alg still achieves OLL; however you will leave 3 corners instead of a G perm.

B R D R' D' B'

Mirroring algs is very important in FMC, they are kind of 'free' and give you many more options.

...

...

Thank you! I hadn't considered B/F mirrors for some reason.

In fact I didn't have a G-perm, but an R-perm. (On that note, is there a better way to deal to insert edge+corner swap PLLs into the skeleton besides a corner 3cycle + an edge 3cycle?)

Also, I have read that blockbuilding from F2L-1 is a very good way to end a skeleton, but I don't really know how it's done.

Thanks again

 

[Brest]

Thank you! I hadn't considered B/F mirrors for some reason.

In fact I didn't have a G-perm, but an R-perm. (On that note, is there a better way to deal to insert edge+corner swap PLLs into the skeleton besides a corner 3cycle + an edge 3cycle?)

Also, I have read that blockbuilding from F2L-1 is a very good way to end a skeleton, but I don't really know how it's done.

Thanks again

Yes, R perm not G perm, although both are two 3 cycles.

Some of the more proficient FMCers can use insertions to solve 2 edges and 2 corners. iirc the idea is the same (stickering & then stepping through the skeleton), but instead of using commutators to cycle 3 pieces you would look for a place where the 4 pieces are positioned correctly for a previously learned algorithm. For example an R perm can be solved in 13 moves, where a J perm can be 10 moves. If you find a place in your skeleton where a J perm would solve the 2E2C then you've saved 3 moves (more if moves cancel). These algorithms are not just PLLs, but rather any short algorithm.

A good way to practice from F2L-1 is to just try as many different ways of pairing then inserting the last pair as you can find. Different insertions R U2 R' instead of R U' R', or the Hammer group R' F R F', to other ideas like using different setup moves F R U2' R' F'. Just experiment with things you know, learn some new tricks, and then try to understand how those algorithms effect the other unsolved pieces. Or just try a bunch of stuff until something works!

 

[Cubenovice]

A good way to practice from F2L-1 is to just try as many different ways of pairing then inserting the last pair as you can find. Different insertions R U2 R' instead of R U' R', or the Hammer group R' F R F', to other ideas like using different setup moves F R U2' R' F'. Just experiment with things you know, learn some new tricks, and then try to understand how those algorithms effect the other unsolved pieces. Or just try a bunch of stuff until something works!

Fully agree, at F2L-1 I always spend some time "just" inserting the pair in different ways and see what OLL's I get, then spam those OLL's with a bunch of short algs and their mirrors. If this leaves me with three corners or edges I continue, otherwise I try some EO / blockbuilding stuff at F2L-1.

Knowing your sunes can be an easy way to get to 3 corners. With 4 edges oriented and two in the right relative positions there's a pretty good chance for leaving three corners. Choose your sune/anti sunes as such that they "swap" the incorrrect edges, do not look at the corners like you would normally do.
For the same LL you can do this with the "bad" edges in 4 orientations using 4 different Sunes; they will all "swap" the edges but give a diffrent end-result.

Sune
Antisune
Sune'
Antisune'

Same goes for LL with two edges oriented, for the L-shape you can do the standard f R U R' U' f' and it's lef thand version but make sure to try some fat sunes from diff angles too.
Only one additional move but you increase you chances of ending in a 3 cycle

O_O 2x2x3??? Is that what FMC is???

No; building a 2x2x3 is just a possible start, nothing more, nothing less

Wouldn't it be good to do 2x2x2 blocks in normal 3x3x3 solving. Sort of like making a reduction method? Because its fast in 4x4x4.

Petrus?
X-cross in CFOP?